Dec., 1951
NOTES
5885
dinitrofluorobenzeneby means of stannous chloride ilinephthalein or 3,3-bis-(p-aminophenyl)-phthaland hydrochloric acid.2 ide. The successful approach to 3,4-difluoroaniline Baeyer2 synthesized this compound from 3,3-bisinvolved the catalytic reduction of 3,4-difluoro- (phenyl)-phthalide and reported it to have a meltnitrobenzene, which was prepared by nitration of ing point of 180'. This synthesis was confirmed by o-difluorobenzene.a The difluoro compound was v. Pechmann,a who, however, did not present any obtained by means of the Schiemann reaction on experimental evidence, and by Fischer and Rome~-,~ o-fluoroaniline, which, in turn, was synthesized who claimed to have obtained dimethylanifrom o-fluorobenzoic acid by means of the Schmidt linephthalein from Baeyer's compound. On the reaction or, less satisfactorily, from o-nitroaniline other hand, Schwarzenbach and Brandenberger,s in a two-step process, (1) the Schiemann reaction, as well as we in our laboratory, failed to obtain the followed by (2) catalytic reduction. Conversion anilinephthalein which melted a t 180' by followof 3,4-difluoroaniline to 3,4-difluorophenylarsonic ing Baeyer's procedure. acid was achieved by means of the Bart reaction. More recently, Schwarzenbach and Brandenberger6 described two new syntheses of anilineExperimental 3,CDitluoroaniline.--Exactly 142 g. (0.89 mole) of 3,4-di- phthalein which they obtained as a compound By their first method, starting fluoronitrobenzene, b.p. 85' (13 mm.), in 150 ml. of absolute melting a t 204'. ethanol was hydrogenated in a Parr apparatus at 60' and 3 from 3,3-bis-(p-tolyl)-phthalide,they obtained it atmospheres pressure in the presence of 2 g. of Raney nickel in very low over-all yields. I n their second and catalyst. Reduction proceeded slowly and fresh catalyst was twice added during the total 48-hour period. The catalyst more efficient method, they condensed phthalyl was then removed by filtration, the solvent by distillation, chloride with s-diphenylurea and then hydrolyzed and the residue distilled under reduced pressure to yield 107 g. the intermediary amorphous condensation product (93%) of 3,4difluoroaniline, b.p. 77" (7 mm.), @D 1.5110. with concd. hydrochloric acid a t 140' in sealed A&. Calcd. for C ~ H ~ N F Z : -55.8; C, H, 3.9; N, 10.9. glass tubes. This method was investigated further Found: (2.55.7; H.3.9; N, 11.0. 3,4-Difluoroaniline hydrochloride was prepared by pass- in our laboratory and an improved procedure was ing dry hydrogen chloride into a solution of the free base in evolved, based on the observation that the condenabsolute ether. The white crystalline product sublimed sation product can be hydrolyzed by a refluxing rapidly above 220'. mixture of acetic and sulfuric acid. A n d . Calcd. for CsH6NFpCl: C , 43.5; H, 3.6; N, 8.5. Several new derivatives of anilinephthalein (I) Found: C, 43.9,43.8; H,3.6, 3.6; N,8.0. 3,4-DilluorophenylarsonicAcid.-A mixture of 1 1. of were prepared, and further proof was obtained that ethanol, 52 g. (0.40) of 3,4difluoroaniline, 40 g. of concen- I actually is the 3,3-bis-(p-aminophenyl)-phthalide. trated sulfuric acid and 112 g. (0.62 mole) of arsenic tri- When reduced, anilinephthalin (11)is formed. On chloride was cooled to o', and 28 g. (0.40 mole) of sodium decarboxylation, this yields 4,4'-diaminotriphenylnitrite in saturated aqueous solution was added with stirring while the temperature was maintained below 5'. When methane (111). In a similar manner, a toluidinephthalein (V) diazotization was complete, 4.0 g. of cuprous bromide was added, the solution was stirred for 30 minutes in the cold, which melted at 197' was prepared from s-di-oand finally heated on a steam-bath until evolution of nitro- tolylurea. It was proved to be 3,3-bis-(p-aminogen ceased. The mixture was steam distilled and the residue evaporated t o 200 ml. and then allowed t o stand for 12 m-tolyl)-phthalide, as follows : By eliminating the hours. The small amount of amorphous material which two amino groups a ditolylphthalide (VI), which precipitated was removed by filtration and discarded. has different properties than those of the known oThe filtrate was concentrated to a volume of 100 ml. and and p-ditolylphthalides, was isolated. Compound cooled; the colorless crystals which formed were washed, first with alcohol and then with ether, and dried to yield 30 VI must, therefore, be the 3,3-bis-(m-tolyl)-phthalg. (31%) of 3,4diiluorophenylarsonic acid, which did not ide, which was further proved by synthesis. '
melt below 300'. An analytical sample was prepared by recrystallization from water. Anal. Calcd. for CsHj0;FaAs: C , 30.3; H , 2.1. Found: C , 30.4; H , 2.3.
Acknowledgment.-These studies were aided by a contract between the Office of Naval Research, Department of Navy, and the University of Kansas. (2) J. J. Blanksma, W. J. Van den Bmek and D. Hoegen, &e. trau. chim., 65, 329 (1946). (3) J. G. McNally and J. R. Byers, U. S. Patent 2,391,179, Dec. 18, 1945.
DEPARTMENT OF CHEMISTRY UNIVERSITY OF KANSAS LAWRENCE, KANSAS
RECEIVED AUGUST9, 1951
Anilinephthalein BY MAXH. HUBACHER
I n the course of a study of compounds with possible laxative properties, it was decided to test an(1) (a) S. Loewe and M. Hubacher, Arch. intern. pharmacodynamic, 65, 297 (1941); (b) M. H. Hubacher, THISJOURNAL, 64,2538 (1942); 66, 25F,(1944): (c) M. H. Hubacher and S. Doernberg. J . A m . Pharm. ASXJL, 37, 261 (1948).
Experimental6 Condensation Product from s-Diphenylurea and Phthalyl Chloride.-A 2-liter flask was charged with 500 ml. of nitrobenzene and 100 g. (0.75 mole) of aluminum chloride (anhydrous, technical or pure grade). After cooling to 25', 40.6 g. (0.2 mole) of phthalyl chloride and finally 42.4 g. (0.2 mole) of sdiphenylurea (m.p. 239-243') were added. The mixture was heated over a period of two hours from room temperature to 80' and kept a t 80' for 30 minutes. The reddish-brown solution was then poured on an ice-hydrochloric acid mixture and the nitrobenzene removed by steam distillation. After the hard, brown residue had been pulverized and submitted to further steam distillation to remove traces of nitrobenzene, it weighed 00-68 g. It starts t o melt at 260" and is not entirely melted a t 300'. It is insoluble in common solvents and could not be obtained in crystalline form. Anilinephthalein (I).-A mixture of 60-08 g. of the condensation product, 400 ml. of acetic acid. 5 ml. of water and (2) A. Baeyer, Ann., POP, 66 (1880). (3) H. v. Pechmann, Bcr., 14, 1866 (1881). (4) 0. Fischer and F. Rbmer, ibid., 42, 2934 (1909). (5) G. Schwarzenbach and M. Brandenberger, Hclo. Chim. Acta. PO, 1253 (1937), and in more detail in "Untersuchungen in der Phtalein-
reihe," Inauguraldissertation von Mark Braadenherger, Univerritat Ziirich, 1936. (6) All the melting points are corrected. Elementary analysis by Oakwold Laboratories, William Saschek and Francine Schwarzkopf.
:iliX(i
NOTES
40 ml. of concd. sulfuric acid was gently refluxed for four hours. Then 70 g. of XatCOI was added to the cooled solution, which exhibited a greenish fluorescence, and the bulk of the acetic acid was distilled off under reduced pressure The gummy residue was dissolved in 600 ml. of boiling 1 A* hydrochloric acid, then 600 ml. of water was added, cooled to 5' and black, tarry matter ieninved by filtration. On making the filtrate ammoniacal, a curdy precipitate was formed, which may agglomerate t o a large lump, gradually becoming hard. The crude I (38.5-46.5 g., m.p. 176186') after one crystallization from yethano1 yielded 2 3 . S 34.9 g. (38-55%) melting above 193 . The pure anilinephthalein, crysttllized from methanol, ( 1 g. in 50 ml.) melts at 202.5-203.2 . It is soluble in acetone, moderately soluble in ethyl acetate (absol.) or methanol, and insoluble in water, benzene or ether (absol.). On addition of I to a warm, concd. alcoholic solution of ZnClz, it slowly turns blue; I also produces the color reaction in acetic acid described by S. and B. The hydrochloride of I, cryst. from concd. HCI, sinters above 220". On coupling diazotized I with an alkaline solution of @-naphthol, an alkali-insoluble tetrazo dye of strong reddish color is formed. Anal. Calcd. for CZOHI~OZN~: C, 75.87; H , 5.09; N, 8.85; mol. wt., 316. Found: C, 76.22; H , 5.32; N, 8.77; mol. wt., 303.1 The formation of impure phenolphthalein from I , as described by s. and B., could be conkmed. On heating the crude phenolphthalein obtained from 0.04 mole of I in a nickel crucible with 100 g. of KOH at 230-235' for ten minutes, 5.33 g. of 4,4'-dihydroxybenzophenone, 3.95 g. of benzoic acid and 0.49 g. of P-hydroxybenzoic acid were formed. anilinephthalin (111.-A solution of 3.16 g. of I in 25 ml. of 5 N hydrochloric acid, to which 5.0 g. of tin (20 mesh) had been added, was refluxed for one day. After adding an additional 25 ml. of acid, refluxing was continued for another day. The decanted solution was made alkaline with 2 A' sodium carbonate (160-172 ml.), the bulk of the solution decanted and the tin hydroxide separated by centrifugation. The last traces of tin were removed as tin sulfide. On addition of 3 N hydrochloric acid to the filtrate to a PH of 6.8 to 7.0, the anilinephthalin precipitated. It was dissolved in 180 to 200 ml. of methanol. Crystals will form after standing for several days at 5' ( I .5 to 1.9 g., m.p. 235237'; yield 49-59 %). It disPure anilinephthaliii (11) melts at 238.1-238.7'. solves either in dil. hydrochloric acid or dil. sodium hydroxide; it is slightly soluble in acetone but insoluble in ether, petroleum ether, ethyl acetate, CCl4, CHCla or benzene. It can be sublimed a t 180' and 5 microns pressure.* 9 solution in concd. sulfuric acid turns greenish on warming. Anal. Calcd. for CzoHlsOaKz: C, 75.45; H, 5.70; X, 8.SO; mol. wt., 318; -COOH, 14.1. Found: C, 75.42; H, 5.36; N, 8.71; mol. wt. 290 (Rast); -COOH, 13.X.9 Decarboxylation of I1 to 4,4'-Diaminotriphenylmethane (111).-The decarboxylation apparatuse was charged with 1.59 g. of 11, 5 ml. of syntheticquinolineand 0.05 g. of basic cupric carbonate. On heating for thirty minutes at 250", 0.2044 g. of CO2 was evolved. After steaming out the quinoline, the dark residue was crystallized from 25 ml. of benzene and the resulting crystals dried a t 120'; yield 1.06 g. (60%,), m.p. 138.3-139.4O.lO The diacetyl derivative, crystallized from a large volume of benzene, melted at 240.0-243.3O and did not depress the n1.p. of an authentic sample of 4,4'-diacetaminotriphenyln1ethane.l' 4,4'-Diacetaminotriphenylmethane-2-carboxylic Acid (IV).-A mixture of 3.18 g. of I1 dnd 10 ml. of acetic anhydride was heated to 110' for 90 minutes. The resulting solution, showing a strong greenish fluorescence, was poured 17) Signer method, as described by E . P. Clark, Ind. Eirg. C h e w , Anal. Ed., 13, 820 (1941). .ketone mas used as a solvent and azubenzene as a standard. (8) M. H. Hubacher, i b i d . , 16, 448 (1943). (9) M. H. Hubacher, Anal. Chem., 41, Y45 (1949). (10) C. Ullmann, J . pvakl. Chcnr., S6, 246 (1887). reported 138.5.139.00. (11) A. Baeyer and V. Villiger, Bcr., 37, 2860 (1904), reported 233234". The 4,4'-diacetaminotriphenylmethaue,prepared according t o their procedure and rrystallized from a ldrge volume of henzene. melted at 243.3-244.1 ".
VOl. 74
011 ice contained in an erleiimeyer flask. The washed, sticky, yellow mass was refluxed for one hour with 40 ml. of 2 N sodium carbonate. The solution was acidified and the yellow precipitate recrystallized from 90 ml. of 41% ethanol; yield 2.9-3.4 g. (73-840/0), m.p. 247-250'. The pure acid (IV), colorless crystals melting at 250.1-251.0', forms a colorless solution in N sodium carbonate. Dissolved in concd. sulfuric acid, the colorless solution turns greenish on heating. Am?. Calcd. for CznHzzO&: C, 71.65; H, 5.51; N, 6.96; mol. wt., 402; -COOH, 11.2. Found: C, 71.11; H, 5.64; N, 7.18; mol. wt. (Rast), 410; -COOH, 10.T9; neut. equiv., 420 i 8. The sodium salt of IV, crystallized from a small quantity of water, was found t o contain 5.170 Na (theory 5.4). The acid was decarboxylated a t 245°,pusing 2.0 g. of IV, 5 ml. of synthetic quinoline and 0.05 g. of basic cupric carbonate. The Droduct melted at 237-242'. after two crvstallizations fro& much benzene.ll The Methyl Ester of 1V.-An ethereal solution of diazomethane, prepared from 3.0 g. of nitrosomethylurea,le was added to a solution of 5.0 g. of IV in 800 ml. of acetone. After evaporation of the solvent the next day, the residue was crystallized from 2 liters of benzene; yield 4.0 g., m.p. 174-180O. The pure ester, crystallized from benzene ( 1 g. in 350 ml.), melts at 180.4182.1°. A n d . Calcd. for Cz~Hz4O&: C, 72.11; H , 5.81; N, 6.73; mol. wt., 416. Found: C, 71.82; H, 5.90; N,7.30; mol. wt., 408.7 o-Toluidinephthalein or 3,3-Bis-(p-amino-m-tolyl)-phthalide (V).-This compound was made by substituting 48.0 g. (0.2 mole) of sdi-o-tolylurea (m.p. 251-253') for the sdiphenylurea in the procedure for the preparation of I. The 76-80 g. of amorphous condensation product yielded This, crystallized from 40-50 g. of crude V, m.p. 180-187'. methanol (1 g. in 50 ml.),gave28.7-38.4g.(42-55'%) melting a t 192-194'. Pure V, colorless needles from methanol, melts a t 196.8197.3'. A 0.1 molar solution in acetic acid turns t o purple on heating to loOD, becoming almost colorless on cooling. Anal. Calcd. for CZZH~OZNZ: C, 76.72; H, 5.85; N, 8.13; mol. wt., 344. Found: C, 76.90; H, 6.12; N,8.59; mol. wt., 349.7 The diacetyl derivative of V, crystallized from methanol (1 gcoin 50 ml.) forms colorless crystals melting a t 256.0256.1 . A n d . Cdlcd. for Cz&2404NZ: c, 72.89; H, 5.64; K, 6.53. Found: C,72.91; H, 5.80; N, 6.64. 3,3-Bis-(m-tolyl)-phthalide (VI). a. From o-Toluidinephthalein.-The two amino groups in V were eliminated by means of the Mai remtion.la A solution of 3.44 g. (0.01 mole) of V in 25 ml. of 2 N hydrochloric acid was diazotized at -4' with a solution of 1.4 g. (0.02 mole) of sodium nitrite in 6 ml. of water. The cold diazo solution was run slowly into 30 ml. of 50% hypophosphorous acid at Oo, contained in a 500-ml. beaker fitted with stirrer. Much foam developed during the reaction. After the temperature was raised overnight to 25O, the amorphous mass was filtered off and dissolved in 10 ml. of ethanol. Reddish crystals separated from t h r cooled solution. On sublimation at 100' and 20 microns pressure,* a snow-white sublimate was obtained. It was purified further, based on the observation that a by-product IS more volatile on sublimation. The pure 3,3-bis-(mtolpl 1-phthalide (VI) forms colorless crystals from methanol and melts a t 112.5-113.2' (1.50-1.72 g., yield 47-55y0). I t is w r y soluble in acetone, carbon tetrachloride or benzene. A 0.01 molar solution in concd. sulfuric acid is strong reddish in color. A n d . Calcd. for C22Hls02: C, 84.05; H , 5.77; mol. w t . , 314. Found: C,84.17; H , 5.83; mol. wt., 32fL7 b. From Phthalic Anhydride and m-1odotoluene.-The Grignard reagent was prepared from 43.6 g. of m-iodotoluene 4.86 g. of magnesium and 400 ml. of ether, the reaction being ___
(12) F. Arndt, "Organic Syntheses." Coll. Vol. 11. John Wiley and Sons, Inc., New York, N. Y . . 1943, p. 165. (13) J. Mai, Bey., 36, 102 (1902); a detailed discussion of this reaction by N. Kornblum can be found in R. Adams, "Organic Reactious," Val. 11, John n'ilep and Sons, Inc., New York, W.Y.. l q 4 4 , pi) 277-282.
NOTES
Dec., 1951
initiated by a few drops of methyl iodide." After the magnesium had dissolved, 160 ml. of benzene was added. Since phthalic anhydride is but slightly soluble in either ether or benzene, 14.8 g. was charged into an extraction thimble, the latter mounted below the reflux condenser. The alkaliinsoluble part was crystallized thrice from ethanol, then sublimed at 100' and 20 microns pressure: yielding 5.67.6 g. melting at 112.0-113.1° (1&24%). A mixture of this compound with that obtained from V melted a t 112113O. Z-(m-Toluyl)-benzoic Acid.-The alkali-soluble portion obtained in the Grignard reaction was purified by sublimation a t 145' and 15 microns pressure,* followed by a crystallization from benzene. The new acid, colorless needles, melts a t 162.2-162.4' (yield 1.6 g.). Anal. Calcd. for C15H1203: C, 75.00; H, 5.04; mol. wt., 240. Found: C, 75.21; H, 4.92; mol. wt., 224'; neut. equiv., 241.5.
5887
be extremely resistant to cleavage and most of the starting material was recovered after the hydrogen chloride treatment. It is evident that in this series, the 2- and 3-isomers react very much like tho 4-is0mer.~
Experimental 3-Bromodibenzothiophepe.-On the basis of a study of several different conditions for the Sandmeyer reartion on 3-aminodibenzothiophene, the general procedure of Hodgson gave the best r e s ~ l t s . ~ A 220-ml. solution of glacial acetic acid containing 19.9 g. (0.1 mole) of the amine' was diazotized by nitrosylsulfuric acid (10% excess) at 15'. The diazonium-salt was added to a warm solution of freshly prepared cuprous bromide (0.1 mole) in 400-ml. of 40% hydrobromic acid solution and the mixture refluxed for one hour. After pouring the cooled liquid into a large volume of water, the product which sepa(14) 0.Grummitt and A. C. Buck, THISJOURNAL, 66, 295 (1943). rated was thoroughly extracted with ether, taken up in hot ethanol, treated with charcoal and after filtration, allowed RESEARCH LABORATORY OF to separate by cooling. The yield of crude product (m.p. Ex-LAX,INC. RECEIVEDMAY4, 1951 93-96') obtained after recrystallization from ethanol was BROOKLYN 17, N. Y. 6 g. (23%). The sample used for analysis melted at 9 7 . 6 98.5'. Anal. Calcd. for CItHTBrS: Br, 30.37; S, 12.18. The Hydrogen Chloride Cleavage of the C-Si Found: Br, 30.52; S , 12.30. Bond in Some Dibenzothienyl-silicon Compounds The structure was proved by preparing the corresponding 5-dioxide, m.p. 224-225', which was identified by compariBY GABRIELLOILLUMINATI, JOHN F. NOBIS AND HENRY son with an authentic specimen.s GILMAN 3-Trimethylsilyldibenzothiophene(II).-To 5.2 8. (0.020 mole) of 3-bromodibenzothioDhene in 50 ml. of dry ether, Recently' 4-trimethylsilyldibenzothiophene (I) under a dry nitrogen atmosphere and a t 5', a titrated soluand its corresponding 5-dioxide (IV) have been tion6 of n-butyllithium (0.022 mole) in ether was first added prepared. It has been observed that, by prolonged and then followed, after 2 minutes, by an ether solution of g. (0.024 mole) of trimethylsilyl chloride. The mixture action of hydrogen chloride in glacial acetic acid, 2.6 was stirred and refluxed for two hours longer, and decom(I) undergoes an extensive cleavage in accordance posed by pouring into water. From the ether layer, a solid with the reaction' residue was obtained, which by crystallization from ethanol yielded 4.1 g. (80%) of white crystalline product melting a t 102-1!4'. The sample used for analysis melted at 103.5104.5 . [Hal (CH&SiCl Anal. Calcd. for C16H&Si: Si, 10.94. Found: Si, 10.74. 2-Trimethylsilyldibenzothiophene (III).-An ether suspension of 8.8 g. (0.035 mole) of 2-bromodibenzothiophene, prepared by direct bromination of diben~othiophene,~ was whereas (IV) remains practically unaffected. allowed to react with n-butyllithium in essential accordance The purpose of the present work was to prepare with the directions described above, and then 4.5 g. (0.042 the 3- and 2-isomers of both (I) and (IV) and to mole) of trimethylsilyl chloride was added. The crude determine the effect of position on ease of cleavage product was distilled a t 159-160" (1 mm.) as a colorless by hydrogen chloride. Both S-trimethylsilyldi- liquid which solidified on standing at 0'; yield 4.5 g. (50%). Further purification by crystallization from ethanol gave benzothiophene (11) and the 2-isomer (111) were white crystals melting a t 48.249.2'. prepared from the respective bromo-derivatives by Anal. Calcd. for C15HleSSi:Si, 10.94. Found: Si, 10.92. halogen-metal interconversion with n-butyllithium 3-Trimethylsilyldibenzothiophene-5-dioxide(V).-A soluand subsequent reaction with trimethylsilyl chlo- tion of 0.004 mole of (11) in 15 ml. of glacial acetic acid was ride. The preparation of 3-bromodibenzothio- treated with 30% hydrogen peroxide (10% excess) and phene was made possible by a recent synthesis of 3- heated between 9610Oo for two hours. When cooled, the solution was poured into water and the resulting solid was aminodibenzothiophene through a rearrangement removed by filtration to give 1.1 g. (98%) of (V), melting reaction of 44ododibenzothiophene with sodamide at 163-170'. Recrystallization from ethanol gave white in liquid ammonia.2 The conversion of the 3- crystals melting at 170.8-171.8'. bromo derivative into (11)was shown to be almost Anal. Calcd. for Cl5HlsOnSSi: Si, 9.73. Found: Si, 9.51. quantitative. 2-Trimethylsilyldibenzothiophene-5-dioxide(VI).-This The corresponding &dioxides (V) and (VI) were was obtained from (111),by using the same proobtained from (11) and (111),respectively, by hy- compound cedure and the same-sized run, described above for the 3; drogen peroxide oxidation. isomer. The yield of the crude product melting a t 157-161
oio
+
The cleavage by hydrogen chloride in glacial acetic acid of (11) and (111) gave results in agreement with that reported for compound (I). That is, the cleavage products identified were dibemothiophene and trimethylsilyl chloride. Furthermore, the 5-dioxides (V) and (VI) were found to
(1) H. Gilmun and J. F. Nobis, THISJOURNAL, 7% 2629 (1950). (2) H. Gilman and J. F. Nobis, ibid., 67, 1479 (1945); see, also, R. K. Brown, R. G Christiansen and R. B. Sandin, i b i d , 70, 1748 (1948).
(3) This illustrates rather strikingly that the position of the so-called negative sulfonyl group with respect to the trimethylsilyl group, in this aromatic system, is without significant influence on the rate of cleavage of the C S i linkage. See, F. C. Whitmore and co-workers, THISJOURNAL, 69, 1551 (1947); L. H. Sommer and co-workers, ibid.. 70, 2869 (1948). (4) H. Hodgson and J. Walker, J. Chcm. SOL, 1620 (1983). (5) H.Gilman and S. Avakian, THIS JOURNAL, 68,1614 (1846). (6) H.Gilman and A. H. Haubein, ibid., 66, 1516 (1944). (7) C. Courtot, L. Nicolas and T. H. Liang, Compr. rand., 186, 1624 (19%); C. Courtot and I. Kelner, ibid., 198, 2003 (1934).
